Network Analysis with Netlytic
Netlytic Workshop at the University of Ottawa with Elizabeth Dubois
Workshop Details:
This workshop introduced participants to the free online tool Netlytic. We reviewed how to set up data collection from various social media sites and collect a small Twitter dataset. Next we reviewed the fundamentals of online social network analysis using Netlytic.
Basics of Social Network Analysis - Online and Offline
What is Social Network Analysis?
Social network analysis helps us figure out relationships between people through their relationships online which helps us link people’s relationships online. People have different ways in which their interactions can be mapped - pointing to what can be learned from their attributes. We are interested in the flow of information and the relationship between different nodes. General idea is to mathematically map out social relationships to determine what the influence is. There are small communities within network graphs. When doing online social network analysis we look at the data left when individuals nagivate the web but we are not able to see what is going on in the real world.
Netlytic
Netlytic is a community-supported text and social networks analyzer that can automatically summarize and discover social networks from online conversations on social media sites. It is made by researchers for researchers, no programming/API skills required.
Tips
- Name your dataset your search term so you do not forget your search term
- Different social networks have different rules
- You can only collect 1000 tweets every fifteen minutes so that servers do not get overloaded with requests
- After the first time you can pull the tweets that occur after you started you will only get what occured within that 15 minutes.
- You can export data into a CSV and open it in Excel to access more data.
- You can also do a text analysis - create wordmaps of what conversations are occuring in the community you are looking at.
- The more ties the denser your charts will be.
- Different layouts are useful for different kinds of graphs - use the one easiest for your to interpret.
- Stars can show those engageing in bot behaviour or acting as promoters.
- You cannot go back in time with Twitter’s data collection unless you pay a data broker.
- When you search a hashtag that is trending you often get a lot of people piggybacking on it.
- A spread out graph can be explained by a sparse network in which people are having these discussions seperately
- Everytime you change your layout your clusters change so if you’re doing research don’t change it a lot and take pictures regularly.
Definitions:
- Node: An individual or user
- Edges (Also called ties):The link between individuals - mentions, retweets, likes, etc.
- Cascade: When one piece of information flows through a network.
- In Degree: How many people engage with it.
- Out Degree: How many people does it engage with.
- Total Degree: All engagement.
- Clusters: Where nodes are grouped together.
- Centrality:
- Betweeness Centality: Connection between different groups.
- Weak Ties: People you are connected to but do not know well. Weak ties are very important because they connect us with different communities.
- Homopholy: Most of the time we connect with people who are just like us.
- Hubs: The key players in a conversation.
- Trace Data: Data we leave as we navigate the web.
- API: A backdoor to the network which researchers can use to pull data.
- Streaming: Opens a door to Twitter and says push through the tweets as they happen.
- Search API: Check in every 15 minutes to allow you to push the data through.